This paper proposes a fractional-order control method for the physical human-robot interaction (pHRI) inspired by the dynamics of muscle contraction-relaxation behavior, which is integrated into a fractional-order sliding mode with variable order and parameters, to achieve fast transient response while guaranteeing high-precision steady-state tracking performance. A sensorless force observer based on fractional calculus is developed to estimate the operator’s behavior, enabling a composite control system that guarantees ultimate boundedness of the closed-loop signals. The deep reinforcement learning-based order and parameter optimization mechanism is synthesized into the unified architecture of the composite control system. The effectiveness of the proposed method is validated through numerical simulations and comparative studies, which demonstrate significant improvements in convergence speed on the sliding manifold while maintaining steady-state precision. Experimental results further confirm the feasibility and practical applicability of the framework in a cylindrical docking scenario via visualreality fusion approach, highlighting its potential for future semiautonomous human-in-loop missions.

This paper primarily introduces a novel prescribed performance control method to achieve rapid and high-precision control in servo systems. Initially, an interesting asymmetric barrier function is proposed, so that the controlled plant with arbitrary initial values can be confined within an asymmetric boundary. To reduce the dependence of controller deployment on physical parameters, an ultralocal model (ULM) approach is adopted and the logarithmic sliding-mode manifold is synthesized to design the controller and observer, resulting in an order-reduced and transient-performance-improved error dynamics. Since there are no non-Lipschitz continuous elements in the logarithmic sliding-mode, the super-twisting algorithm can be used to weaken signal chattering while converging the equivalent error rapidly. The Lyapunov-based direct analysis proves the stability of the controlled servo system with unknown parameters. The superiority of the scheme is verified through a series of simulations and experiments on PMSM platform.

Autonomous fine manipulation in space for orbital assembly continues to present a critical challenge in the field of aerospace engineering. Under low-gravity conditions, during satellite manipulator operations on free-floating objects, the absence of significant gravitational forces and friction constraints leads to unpredictable relative motions between the manipulator’s end-effectors and objects, degrading the manipulation performance. This study proposes a fine manipulation method for satellite robots with floating platforms, grounded in multimodal perception and an enhanced Action Chunking with Transformer (ACT) architecture that enables multimodal state interaction. By integrating visual, tactile, and depth sensory data, the satellite robot’s space fine manipulation capabilities are substantially improved. From the experimental results generated from simulated environments, the proposed method achieves a success rate exceeding 80% for peg-in-socket insertion tasks, outperforming conventional approaches with a success rate of approximately 45%. Project Website: https://github.com/LSY0528/VDTF-ACT.

This article addresses the suboptimal high-precision tracking problem for a class of nonlinear systems, considering constraints on both state and input, by employing an asymmetric prescribed performance envelope. To make the equilibrium of the controlled plant coincide with its equivalent, a novel error transformation is proposed in this paper so that the equivalent tracking error can be mapped to an asymmetric prescribed boundary containing the origin. To address the optimal tracking control issue of the saturated nonlinear system with the non-zero equilibrium, a modified cost function corresponding to equivalent asymmetric state transformation is proposed, simplifying designing a saturated ADP error tracking controller. Since the equivalent system with asymmetric state constraints can be previously reduced to a logarithmic sliding-mode system, a fine-tracking precision is achieved by a high gain at the equilibrium while reducing the implementation difficulty of the proposed ADP controller. Experiments conducted on a permanent magnet synchronous motor (PMSM) system verify the effectiveness of the proposed control scheme.

Path Integral Strategy Improvement (PI2)-based impedance control is a superior scheme for preventing damage to the physical structure of the fruit during the harvesting process. However, it is highly sensitive to disturbances and has limited generalization ability during the parameter learning process, making it difficult to apply the correct gripping force to fruits with uncertain stiffness. To solve this problem, this paper proposes a variable impedance force control method that integrates PI2 with Dynamic Movement Primitives (DMPs), supplemented by a force error correction loop. Firstly, an adaptive impedance parameter matching mechanism based on gain schedules is designed to facilitate dynamic estimation of impedance parameters and enable precise force control. To further accelerate impedance parameter matching in unknown environments, the PI2 algorithm is introduced to optimize gain schedules, and DMPs are integrated to suppress disturbances, thereby improving the generalization ability of the impedance model’s parameter learning. In addition, an additional force error control loop has been designed to minimize the deviation between the desired and actual gripping force. Finally, the effectiveness of the proposed method is verified through simulation and experiment in fruit teleoperation picking robot.

This article addresses the zonotope-based ${\mathcal {L}}_{\infty }$ event-triggered control for switched positive systems with frequent asynchronism and interval uncertainties. In view of the advantages of zonotope in set description, we extend it to the set-based control, propose a novel zonotope-based control method, and further consider the typical model uncertainty and frequent asynchronism. First, by introducing a 1-norm-based event-triggered scheme, the time-varying state zonotope is established based on the closed-loop system and event-triggering conditions. Second, the dual convergence and ${\mathcal {L}}_{\infty }$ performance of the positive state zonotope is analyzed by defining suitable center and radius functions for matched and mismatched intervals, respectively. Then, the permissible mode-dependent average dwell time signals and asynchronous controllers are jointly designed to guarantee the stability and ${\mathcal {L}}_{\infty }$ performance for the underlying system. Finally, the effectiveness and advantages of the proposed method are demonstrated through a numerical example.

In this article, the event-triggered fixed-time tracking control is investigated for uncertain strict-feedback nonlinear systems involving state constraints. By employing the universal transformed function (UTF) and coordinate transformation techniques into backstepping design procedure, the proposed control scheme ensures that all states are constrained within the time-varying asymmetric boundaries, and meanwhile, the undesired feasibility condition existing in other constrained controllers can be removed elegantly. Different from the existing static event-triggered mechanism, a dynamic event-triggered mechanism (DETM) is devised via constructing a novel dynamic function, so that the communication burden from the controller to actuator is further alleviated. Furthermore, with the aid of adaptive neural network (NN) technique and generalized first-order filter, together with Lyapunov theory, it is proved that the states of closed-loop system converge to small regions around zero with fixed-time convergence rate. The simulation results confirm the benefits of developed scheme.

This article considers the implementation of an observer-based logarithmic control scheme for physical human-robot interaction, which is a typical Lagrangian system. The novelty lies in using a switching term in the logarithmic sliding mode observer to describe the operator’s behavior without any sensors, and applying adaptive parameters in the logarithmic sliding mode controller (SMC) to practically stabilize reaching the sliding surface using chattering-free nonsingular reaching law in finite time. A reset mechanism is synthesized into the control system to enhance the transient response. The motion on the sliding surface is analyzed from the perspective of practical finite-time stability, from which both the convergence regions of the tracking and estimate errors can be determined. The numerical and experimental results verify the effectiveness and advantage of the proposed reset logarithmic SMC and observer for human-robot interaction compared to the existing linear SMC and terminal SMC. With regards to settling time and rising time, the superiority of transient performance in experimental results is coincident with the stability analysis.

This paper develops a disturbance observer-based discrete-time sliding mode control (DOB-DSMC) method to address the trajectory tracking problem for nonholonomic robots subject to disturbances. Firstly, a reduced-dimensional tracking error system is derived to overcome the disadvantage of the underactuated robot system. In the light of reduced-dimensional system, a disturbance observer is studied to estimate and compensate for disturbances. Then, inspired by the continuous terminal sliding mode surface, a novel discrete-time terminal sliding surface is designed. Further, the arctangent function is inducted to improve the tracking accuracy and attenuate the chattering phenomenon. Moreover, the asymptotic stability of system is proved via the Lyapunov method. Finally, the effectiveness of the exploited algorithm is verified by a numerical simulation.

The cross-domain unmanned system represents a system that interacts with information between the unmanned ground vehicle (UGV) and the unmanned aerial vehicle (UAV) to accomplish cooperative behavior to dominate a span of air-ground domains. This paper presents a behavior-aware co-operative scheme based on backstepping control and master-slave structure with an admittance mechanism. According to the dynamics of cross-domain unmanned systems, backstepping controllers are designed for UAV and UGV, respectively, and the Lyapunov stability analyses ensure the theoretical feasibility. The master-slave structure-based cooperative scheme is presented, where the behavior-aware cooperative strategy is generated by an admittance mechanism to promote the leader's understanding of the follower's cooperation. The simulation results illustrate that the proposed cooperative scheme is effective in traceability and robustness with disturbances.

Reinforcement learning (RL) is an effective approach to motion planning in autonomous driving, where an optimal driving policy can be automatically learned using the interaction data with the environment. Nevertheless, the reward function for an RL agent, which is significant to its performance, is challenging to determine. The conventional work mainly focuses on rewarding safe driving states but does not incorporate the awareness of risky driving behaviors of the vehicles. In this paper, we investigate how to use risk-aware reward shaping to leverage the training and test performance of RL agents in autonomous driving. Based on the essential requirements that prescribe the safety specifications for general autonomous driving in practice, we propose additional reshaped reward terms that encourage exploration and penalize risky driving behaviors. A simulation study in OpenAI Gym indicates the advantage of risk-aware reward shaping for various RL agents. Also, we point out that proximal policy optimization (PPO) is likely to be the best RL method that works with risk-aware reward shaping.

This paper presents a dynamic event-triggered communication mechanism to mitigate limited communication in multi-UAV systems. This mechanism aims to reduce resource consumption within these systems by enabling UAVs to communicate only when specific triggering conditions are met. Unlike static event-triggered communication mechanisms, our approach incorporates a dynamically adjusted triggering threshold through a carefully designed dynamic rule. As a result, the system performance is enhanced while communication resource utilization in multi-UAV systems is reduced. Building upon this dynamic event-triggered communication mechanism, we propose a distributed formation control strategy. Furthermore, we outline a criteria for designing the relevant control parameters. To validate the proposed approach, numerical simulation is conducted.

This article investigates the stabilization of the bilateral teleoperation system with inherent strong nonlinearities and model parametric uncertainties, and a fractional-order non-singular terminal sliding mode control (FONTSMC) strategy for high-precision master-slave tracking is generated by combining fractional calculus and sliding mode control. The stability of the closed-loop system and finite-time convergence are analyzed using the Lyapunov stability theory. For simulation verification, the presented control scheme is applied to a pair of 3-DoF Phantom Omni haptic manipulators. The results reveal that the performance of non-singular terminal sliding mode control with fractional calculus is significantly better than that of the integer-order non-singular terminal sliding mode control in terms of trajectory tracking accuracy and response time.

In this paper, the control performance of the teleoperation system subjected to dynamics uncertainty and external disturbance is investigated. To improve control performance, an adaptive neural learning prescribed-time controller was developed, which ensures that the system’s output tracks the desired trajectory with a predetermined accuracy within a user-defined time. Unlike other general finite-time or fixed-time controllers, the predetermined convergence time can be exactly obtained rather than approximated. Moreover, the proposed control scheme can solve the issue with and without constraints uniformly. With the aid of the Lyapunov method, the stability of the system is analyzed. Finally, the effectiveness of the proposed method is further verified by numerical simulations.

This article proposes an adaptive neural-network control scheme for a rigid manipulator with input saturation, full-order state constraint, and unmodeled dynamics. An adaptive law is presented to reduce the adverse effect arising from input saturation based on a multiply operation solution, and the adaptive law is capable of converging to the specified ratio of the desired input to the saturation boundary while the closed-loop system stabilizes. The neural network is implemented to approximate the unmodeled dynamics. Moreover, the barrier Lyapunov function methodology is utilized to guarantee the assumption that the control system works to constrain the input and full-order states. It is proved that all states of the closed-loop system are uniformly ultimately bounded with the presented constraints under input saturation. Simulation results verify the stability analyses on input saturation and full-order state constraint, which are coincident with the preset boundaries.

This article proposes a hybrid control scheme synthesizing fuzzy approximate Q-iteration algorithm and discrete-time terminal-like sliding mode control for deploying tethered space robot, which is modeled as a deterministic Markov decision process. The existence of a switching condition allows FQ-iteration algorithm and terminal-like sliding surface constituting an optimal sliding mode control, and the fuzzy logic approximation is employed to improve the efficiency of optimization. Under arbitrary switching, the sliding mode reaching law works to compress the contraction of sliding surface variable. Simulation results verify the analyses on contraction of fuzzy approximate Q-iteration for optimal sliding mode control, the stability of reduced-order system yielded by the proposed discrete-time terminal-like sliding surface, and existence of switching condition.

A novel full-order sliding mode tension control scheme for the deployment/retrieval of the space tether system is proposed. The deployment/retrieval dynamics of the space tether system are derived by using Lagrangian mechanics theory. The ideal full-order sliding mode surfaces of the deployment/retrieval dynamics are design using KTC and the second method of Lyapunov, and the designed control technologies can guarantee the asymptotic stability of the full-order sliding mode dynamics. The continuous input is applied to ensure that the system states can reach the ideal surfaces in finite time and keep stable in the subsequent time. The positive tension limit is taken into consideration with choosing appropriate parameters or gains in the design of the full-order sliding mode controller. The numerical results valid the effectiveness of the proposed methods.